a massive spiral galaxy in the zone of avoidancemensa.ast.uct.ac.za/~kraan/papers/06massive.pdf ·...

Mon. Not. R. Astron. Soc. 000, 000–000 (0000) Printed 4 April 2006 (MN LATEX style file v2.2)

A Massive Spiral Galaxy in the Zone of Avoidance?

J. L. Donley1,2, B. S. Koribalski1, L. Staveley-Smith1, R. C. Kraan-Korteweg3,4,

A. Schroder5, P. A. Henning61Australia Telescope National Facility, CSIRO, P.O. Box 76, Epping, NSW 1710, Australia2Steward Observatory, University of Arizona, 933 N. Cherry Ave., Tucson, AZ 85721, USA3Department of Astronomy, University of Cape Town, Private Bag X3, Rondebosch, 7701, South Africa4Depto. de Astronomıa, Universidad de Guanajuato, Apdo. Postal 144, Guanajuato, GTO 36000, Mexico5Department of Physics and Astronomy, University of Leicester, Leicester LE1 7RH, UK6Institute for Astrophysics, University of New Mexico, 800 Yale Blvd., NE, Albuquerque, NM 87131, USA

Accepted 2006 April 4. Received 2006 March 31; in original form 2005 Dec 7

ABSTRACT

We report the discovery of a very H i-massive disk galaxy, HIZOAJ0836–43, at avelocity of vhel = 10689 km s−1, corresponding to a distance of 148 Mpc (assumingH0 = 75 km s−1 Mpc−1). It was found during the course of a systematic H i surveyof the southern Zone of Avoidance (|b| ≤ 5) with the multibeam system at the 64mParkes radio telescope. Follow-up observations with the Australia Telescope CompactArray (ATCA) reveal an extended H i disk. We derive an H i mass of 7.5 × 1010 M.Using the H i radius, we estimate a total dynamical mass of 1.4 × 1012 M, similarto the most massive known disk galaxies such as Malin 1. HIZOA J0836–43 lies deepin the Zone of Avoidance (`, b = 262.48,−1.64) where the optical extinction is veryhigh, AB = 9.m8. However, in the near-infrared wavebands, where the extinction isconsiderably lower, HIZOAJ0836–43 is clearly detected by both DENIS and 2MASS.Deep AAT near-infrared (Ks and H-band) images show that HIZOAJ0836–43 is aninclined disk galaxy with a prominent bulge (scale length 2.5′′ or 1.7 kpc), and anextended disk (scale length 7′′ or 4.7 kpc) which can be traced along the major axisout to a radius of 20′′ or 13.4 kpc (at 20 magarcsec−2 in Ks). The H i disk is much moreextended, having a radius of 66 kpc at 1 M pc−2. Detections in the radio continuumat 1.4 GHz and at 60 µm (IRAS) are consistent with HIZOAJ0836–43 forming starsat a rate of ∼35 M yr−1. We compare the properties of HIZOA J0836–43 with thoseof the most H i-massive galaxies currently known, UGC 4288, UGC 1752 and Malin 1,all of which are classified as giant low surface brightness galaxies.

Key words: galaxies: individual (HIZOA J0836–43) — galaxies: formation

1 INTRODUCTION

Hierarchical models of structure formation in the early Uni-verse predict that massive galaxies begin to form from themerging of low-mass systems at early times and that themass and luminosity functions for galaxies evolve in a self-similar manner (Press & Schechter 1974; Cole et al. 2000).Such models (Kauffmann, White & Guiderdoni 1993; Klypinet al. 1997) predict large numbers of old low-mass systemsand relatively small numbers of young high-mass systems.

? The observations were obtained with the Australia Telescopewhich is funded by the Commonwealth of Australia for operationsas a National Facility managed by CSIRO.

However, the detailed match between observations and pre-dictions of the mass function is poor. Two well-documenteddiscrepancies are (1) the small number of low-mass systemsobserved in the local Universe (Davies, Sabatini & Roberts2004) and (2) the large number of massive, distant red galax-ies, implying a significant density of massive galaxies at earlytimes (Foerster Schreiber et al. 2004).

These discrepancies most probably arise because of thefundamental role of gas physics (supernova and AGN feed-back, etc.) in determining the star-formation efficiency inhalos of different mass (Kay et al. 2002). Indeed, a recentattempt to incorporate gas physics by the application of asuite of semi-analytic models to the Millennium run (Crotonet al. 2006, Springel et al. 2005) has been remarkably suc-

c© 0000 RAS

2 J. L. Donley et al.

Figure 1. Global H i profile of the galaxy HIZOA J0836–43 as obtained with the 64-m Parkes telescope (thick line) and the ATCA750D-array (thin line).

cessful in reproducing the exponentially decreasing volumedensity of objects more luminous than a characteristic valueL∗, as well as the observation that massive galaxies tend tocontain older stars than low-mass systems.

Due to the sharp decline in the number of galaxieswith L > L∗ (or M > M∗), large samples are needed toconstrain the high-luminosity and high-mass ends of thegalaxy luminosity and mass functions. In the optical, sucha sample is provided by the 2dF Galaxy Redshift Sur-vey, from which Norberg et al. (2002) find M∗

bj= −19.m0

(H0 = 75 kms−1 Mpc−1) and a slight tendency for an over-abundance of massive galaxies relative to a Schechter fit (al-though they comment that magnitude errors are mostly ableto account for this deviation). Similar results are obtainedfrom the Sloan Digital Sky Survey (Blanton et al. 2003).

To sample the high-mass end of the H i mass function,blind H i surveys such as the H i Parkes All-Sky Survey(HIPASS) are required. HIPASS is a systematic H i sur-vey of the southern sky within the radial velocity range of−1200 < cz < 12700 kms−1 (Meyer et al. 2004). Zwaan etal. (2005) find a characteristic mass in neutral hydrogen gasof M∗

HI = 6.3 × 109 M (H0 = 75 kms−1 Mpc−1), using acatalogue of 4315 H i galaxies from HIPASS. The mass func-tion remains ill-constrained at the high-mass end, however,due to the increasing sparsity of high H i-massive galaxies.

We report here on the discovery and detailed proper-ties of the most massive galaxy in the H i Zone of AvoidanceSurvey (HIZOA), a deep Parkes H i multibeam survey in theobscured region along the Galactic Plane (§2) (see Donley etal. 2005, Kraan-Korteweg et al. 2005, and Henning, Kraan-Korteweg, & Staveley-Smith 2005 for preliminary results).This galaxy, HIZOAJ0836–43, is also the most H i-massivegalaxy in the southern and northern-extension HIPASS cat-alogues, which consist of 4315 and 1003 H i sources, respec-tively (Meyer et al. 2004, Wong et al. 2005). At the mass ofHIZOAJ0836–43, MHI = 7.5 × 1010 M, the volume den-sity is 1.4×10−9 Mpc−3 dex−1, a factor of 106 times smallerthan the corresponding value at M∗

HI.

A study of massive gaseous (and stellar) systems is ofconsiderable interest in assessing whether the details of vari-ous galaxy formation scenarios, particularly the processes ofmerging and star-formation, remain viable under the mostextreme conditions. The H i-massive galaxy to have received

most attention is Malin 1, whose H i mass is reported to besomewhere in the range 5−10×1010 M (Bothun et al. 1987;Pickering et al. 1997; Matthews et al. 2001)1 . Malin 1’s highH i mass but low optical surface brightness is of interest inthat it raises the possibility of a cosmologically significantamount of gas being locked up in such systems. However,such a population is not known to exist at the sensitivitiesof existing H i surveys (Rosenberg & Schneider 2002; Zwaanet al. 2003).

The radio and infrared properties of the newly-discovered galaxy are detailed in §3. In §4, we look at the H i

and dynamical mass of HIZOAJ0836–43, its star-formationrate, morphology, location on the Tully-Fisher relation, andenvironment. A comparison with other H i-massive galaxiesis given in §5, followed by our conclusions in §6.

2 DISCOVERY OF HIZOA J0836–43 WITH

THE PARKES TELESCOPE

The galaxy HIZOAJ0836–43 was first detected in the deepH i Zone of Avoidance (ZOA) survey made with the Parkes64-m telescope between March 1997 and June 2002 (seeKraan-Korteweg et al. 2005). This blind survey utilised theParkes 21-cm multibeam receiver, an array of 13 beams eachwith two orthogonal linear polarisations (for a detailed de-scription see Staveley-Smith et al. 1996). The data werecalibrated and gridded using the aips++ programs live-

data and gridzilla, respectively; after gridding the beamwidth is 15.′5. The data reduction is nearly identical tothat of HIPASS (see Barnes et al. 2001, Henning et al.in preparation). The H i Zone of Avoidance survey coversthe Galactic longitude range 52 < ` < 196, compris-ing the southern Milky Way (Henning et al. in prepara-tion) plus a northern extension (Donley et al. 2005), forthe most opaque part of the ZOA, i.e. the latitude range−5 < b < 5. With a correlator bandwidth of 64 MHzdivided into 1024 channels and an instantaneous velocity

1 Calibration uncertainties at the redshift of Malin 1 seem tobe responsible for the large uncertainties in published massestimates.

c© 0000 RAS, MNRAS 000, 000–000

A Massive Spiral Galaxy in the Zone of Avoidance 3

Figure 2. H i distribution (left) and mean H i velocity field (right) of the galaxy HIZOA J0836–43. Contour levels are 0.1, 0.3, 0.6, 1,1.5, 2.0, 2.5, and 3.0 Jy beam−1 km s−1 (top left; 1 Jy beam−1 km s−1 equals an H i column density of 6.8 × 1020 cm−2), 0.1, 0.2, 0.4,0.8, 1.2, and 1.6 Jy beam−1 km s−1 (bottom left; 1 Jy beam−1 km s−1 equals an H i column density of 3.0 × 1021 cm−2), and 10450 to10950 km s−1 in steps of 50 km s−1 (right). The synthesized beam is shown in the bottom left corner of each frame (47.5′′ × 36.7′′ onthe top, and 20′′ on the bottom). The cross marks the galaxy centre as determined from the 20-cm radio continuum (see Table 2).

coverage of −1200 < cz < 12700 kms−1, the velocity spac-ing per channel is 13.2 kms−1, or 26 kms−1 after Hanningsmoothing. The approximate total integration time is 2100sec beam−1, resulting in an rms of 6 mJy beam−1 in largeparts of the survey. Results of a shallow subset with an rmsof 15 mJy beam−1, the ZOA H i-shallow survey (HIZSS), aregiven in Henning et al. (2000).

While visually inspecting the data cube centred on` = 264, an exceptionally strong signal (Speak = 47mJy beam−1; see Fig. 1) for its very high velocity (vhel ≈10700 kms−1) was initially identified at α, δ(J2000) =08h 36.9m, –43 38′ (±3′) or ` = 262.5, b = −1.6. The de-tected H i source, named HIZOAJ0836–43, remains unre-solved in the Parkes data.

As seen in Fig. 1, the H i signal of HIZOAJ0836–43 is exceptionally broad with a velocity width of ∆v ≈600 kms−1. Using the Parkes data we measure an H i fluxdensity of 13.2 Jy km s−1 corresponding to an H i mass ofMHI = 7 × 1010 M (assuming D = 148 Mpc, see Sec-tion 3.1.1). This same signal was later independently identi-fied in the recently released H i catalogue (HICAT, Meyer etal. 2004) obtained from HIPASS. It is listed there as HIPASSJ0836–43. The H i parameters extracted from the respectivesurveys are given in Table 1.

Detailed investigation of HIZOAJ0836–43 is extremelydifficult due to its location behind the Milky Way. Accord-ing to the DIRBE/IRAS dust extinction maps of Schlegel

et al. (1998) the B-band extinction at the position of HI-ZOAJ0836–43 is AB = 9.m8, although we caution thatthe data in the Galactic plane are not yet well calibrated.Not unexpectedly, this galaxy has no known optical coun-terpart, and inspection of the DSS images reveal only avery faint R-band counterpart. The corresponding extinc-tion in the near-infrared (NIR) I, J , H, and K bandsis much lower: AI = 4.m4, AJ = 2.m2, AH = 1.m3, andAK = 0.m8, respectively. DENIS (IJK; Epchtein 1997) and2MASS (JHK; Jarrett et al. 2000) reveal two and threepossible counterparts, respectively, in all the wavebands,2MASX J08365157–4337407, 2MASX J08370723–4339137and 2MASX J08363600–4337556. As discussed in §3.1.1and §3.2, the counterpart to HIZOAJ0836–43 is 2MASXJ08365157–4337407, a galaxy with an inclination of ∼ 66.

3 RADIO AND INFRARED RESULTS

3.1 ATCA radio observations

H i line and 20-cm radio continuum observations of thegalaxy HIZOAJ0836–43 were obtained with the AustraliaTelescope Compact Array (ATCA) in the 750D and the1.5D configurations. The observations took place on 2003February 10 and November 28, respectively, with a totalintegration time of ∼ 2× 12 hours. For the H i line observa-tions, we used a centre frequency of 1371 MHz and a band-

c© 0000 RAS, MNRAS 000, 000–000


Figure 3. Top: H i distribution of HIZOA J0836–43 (contours) at a resolution of 20′′ overlaid onto the AAT Ks-band image (greyscale).The contour levels are the same as in Fig. 2 (bottom left). Bottom: 20-cm radio continuum image of HIZOA J0836–43 (contours)overlaid onto the same K-band image. The contour levels are –0.26, 0.26, 0.52, 1.04, 2.08, and 4.16 mJy beam−1; the synthesized beamis 8.7′′ × 6.5′′. The displayed region was chosen to also include the two neighbouring galaxies, one of which is detected in the radiocontinuum with an H i flux density of ∼ 1 mJy.

width of 8 MHz, divided into 512 channels. This resulted ina channel width of 3.4 kms−1 and a velocity resolution of4.1 km s−1. The wide-band 20-cm radio continuum observa-tions were centred on 1384 MHz with a bandwidth of 128MHz, divided into 32 channels. We used PKS B1934–638 asprimary calibrator and PKS B0823–500 as secondary cali-brator. Their flux densities are 15.0 Jy and 5.5 Jy at 1378MHz, respectively. Data reduction was carried out with themiriad software package using standard procedures.

Various H i line data cubes were made to investigate theneutral hydrogen distribution and kinematics of the galaxyHIZOAJ0836–43. The best results were obtained by us-ing ‘natural’ weighting of the combined uv-data sets. In-cluding all baselines resulted in an angular resolution of

18.9′′ × 18.8′′ ; we used a restoring beam of 20′′ and a chan-nel width of 10 kms−1. By excluding the longest baselines(i.e. all baselines to antenna 6) we obtain an angular res-olution of 47.5′′ × 36.7′′. We measure an rms per channelof 1.2 mJy beam−1 and 1.4 mJy beam−1, respectively. Theresulting H i distribution and mean H i velocity field of HI-ZOAJ0836–43 for both angular resolutions are shown inFig. 2.

Radio continuum images were made using ‘robust’ (r =0) and ‘uniform’ weighting of the combined uv-data, result-ing in angular resolutions of 10.3′′ × 9.0′′ and 8.7′′ × 6.5′′,respectively. The measured rms is ∼0.1 mJy beam−1. Wefind a continuum source at the position of HIZOAJ0836–43that is slightly extended along a position angle of PA =

c© 0000 RAS, MNRAS 000, 000–000


Figure 4. Radial H i density profile of HIZOA J0836–43, obtainedwith the program radial in the software package gipsy. The solidline (and filled circles) gives the average profile, whereas dottedand dashed lines give the H i profile of the approaching and re-ceding sides, respectively.

125 and that has an integrated 20-cm flux density of ∼24mJy.

3.1.1 Global HI properties

The global H i spectra of HIZOAJ0836–43 (see Fig. 1) wereanalysed using the miriad program mbspect. Because thegalaxy was unresolved in the Parkes H i data, its spectralprofile was formed by weighting the flux according to thebeam parameters. In the ATCA H i data, HIZOAJ0836–43is clearly resolved and consequently the spectral profile wastaken to be the sum of the flux in a conservative box of 3.′8×2.′2, centred on the galaxy. For both the Parkes and ATCAH i data, first-order baselines were fit to the spectra and thetotal H i flux density was then determined by integratingover the galaxy profile.

The systemic velocity was taken to be the midpoint ofthe spectral profile at the 50% H i peak flux level. From aheliocentric velocity of vhel = 10689 km s−1 (z = 0.036), wecalculate a velocity in the Cosmic Microwave Background(CMB) frame of vCMB = 10928 km s−1, where the CMB vec-tor was taken from Fixsen et al. (1996). Adopting a Hubbleconstant of H0 = 75 kms−1 Mpc−1 and a cosmology withΩm = 0.3 and Ωλ = 0.7, we derive a cosmological luminositydistance of D = 148 Mpc (see Table 2). At this distance, 1′

corresponds to 40.1 kpc.The 20% and 50% velocity widths were measured using

a width-maximizing technique. The measured H i proper-ties for both observations are given in Table 1, where theyare compared with equivalent measurements from HICAT(Meyer et al. 2004). For the remainder of the paper, weadopt the ATCA velocity widths of w20 = 610 ± 9 km s−1

and w50 = 566 ± 6 km s−1. Uncertainties in the integrated

flux density and velocity measurements were derived usingthe formulas given in Koribalski et al. (2004).

Fig. 3 shows the Ks-band image obtained with theAAT (see §3.2 for further details) overlaid onto the H i dis-tribution (upper panel) and continuum distribution (lowerpanel) as obtained from the high-resolution ATCA H i datacube. Both the H i and the 20 cm continuum measure-ments identify the galaxy 2MASX J08365157–4337407 asthe counterpart to HIZOAJ0836–43. The other two galax-ies, 2MASX J08363600–4337556 and 2MASX J08370723–4339137, lie at distances of 2.′8 and 3.′2 from HIZOAJ0836–43, respectively.

3.1.2 HI radial density profile

We measure the H i radial density profile from the H i to-tal intensity map of HIZOAJ0836–43, using the iterativemethod of Lucy (1974) as applied by Warmels (1988) andimplemented in the program radial in the software pack-age gipsy (van der Hulst et al. 1992). Because this methodassumes that the H i has a planar and axisymmetric distri-bution, it is limited in its detection of structures such asspiral arms and warps. However, for highly-inclined galaxiesobserved with low angular resolution, such as HIZOAJ0836–43, this method provides the most accurate means of mea-suring the surface density distribution. Another advantageof the Lucy method is that the surface density is measureddirectly; we need not assume a given model to which thedata are fit. In addition, the density profile can be measuredindependently for the two sides of the galaxy.

The H i radial density profile for HIZOAJ0836–43 as de-rived from the low resolution H i distribution (Fig. 2, left) isgiven in Fig. 4. The density profiles of both the approachingand receding sides of the galaxy are shown, as is the averagedistribution. The centre of the galaxy was taken to be theposition of the radio continuum source (see Table 2). Theradial density profile shows a central depression, and peaksat a surface density of 8.6 M pc−2 and 9.1 M pc−2 at radiiof 25 kpc and 45 kpc, on the approaching and receding sidesof the galaxy, respectively. The H i radius, measured at theposition where the average radial density of the H i profiledrops to 1 M pc−2, is ∼66 kpc.

3.1.3 HI rotation curve

The H i rotation curve was measured using the envelope-tracing method (Sofue 1996, 1997), a method known toprovide reliable rotation curves for highly inclined galaxies(Olling 1996). The rotation velocity is given by:

Vrot = (Vt − Vsys)/sini − (σ2obs + σ2

ISM)1/2 (1)

where σobs is the velocity resolution and σISM is the velocitydispersion of the interstellar gas. We adopt a velocity disper-sion of 7 km s−1 (Stark & Brand 1989; Malhotra 1994). Theterminal velocity, Vt, is the velocity at which the intensitybecomes:

It = [(0.2Imax)2 + I2

3σ]1/2 (2)

where Imax and I3σ are the maximum intensity and 3σ in-tensity levels, respectively.

c© 0000 RAS, MNRAS 000, 000–000


Figure 5. H i position-velocity (pv) diagram along the major axis (PA = 285) of HIZOA J0836–43 using the combined H i data setwith an angular resolution of 20′′ . The white line gives the H i rotation curve as shown in Fig. 8. Contours represent intensity levels of2–8σ, where σ = 1.14 mJy beam−1, and are measured after smoothing the pv-diagram by a Gaussian kernel of width 2.

Figure 6. Sample H i velocity profiles of HIZOA J0836–43 obtained from the major-axis pv-diagram at offsets of –0.9′ and 1.0′. Dottedlines represent the rotation velocity measured for each slice. Both the unsmoothed and the Savitzky-Golay smoothed profiles are shownfor comparison.

To measure Vt, we first used the karma package kpvs-

lice to create a position-velocity (pv) diagram along themajor axis of the galaxy (using PA = 285), shown in Fig. 5.We then plotted as a function of velocity the intensity at agiven radius along the major axis. Because of the relativelylow S/N, we smoothed the profiles using a Savitzky-Golaysmoothing filter with a width of 10 pixels (Press et al. 1992)before measuring Imax and the velocity at which I = It.Fig. 6 illustrates the effect of this smoothing on two of thevelocity profiles, as well as the resulting rotation velocitymeasured for each slice. After measuring the rotation veloc-ity at each offset, we smoothed the rotation curve by thebeam size. Because the image resolution limits the reliabil-ity of the central rotation curve derived using the envelope-tracing method (Sofue 1996), we excluded from our rotationcurve measurement a central region equal to the beam di-ameter. We then connected the two sides of the curve with a

line, although we caution that this shape is not necessarilyrepresentative of the true central rotation curve.

The inclination-corrected rotation curve is shown inFig. 7; it is also overlaid onto the pv-diagram shown in Fig. 5.The rotation velocity reaches a local maximum at a radiusof ∼15 kpc, drops slightly, and then rises to the inclination-corrected value of 304 km s−1, remaining relatively flat andsymmetric out to a radius of ∼50 kpc. The rotation velocityis in close agreement with that implied by the inclination-corrected 50% velocity width (310 kms−1).

3.2 Infrared observations and photometry

The 4-m Anglo-Australian Telescope (AAT) was used to ob-tain Ks- and H-band images of HIZOAJ0836–43 on 2003April 9, with exposure times of 660 s and 420 s, respectively,and 1′′.8−2′′.0 seeing. The final images were made by randomdithering of 11 Ks-band images and 7 H-band images in a

c© 0000 RAS, MNRAS 000, 000–000


Figure 7. H i rotation curve of HIZOA J0836–43 derived usingthe envelope-tracing method.

±30′′ box. The positions were calibrated using the koords

tool in Karma (Gooch 1996), with a DENIS (Epchtein 1997)image of the same field serving as the reference. The DENISpositions are accurate to better than 1′′.

To obtain accurate photometry of the three galaxies inthe crowded fields, we first measured and subtracted all ofthe stars in the field using the IRAF2 package killall (seeWoudt et al. 2005 for a detailed description). The resultingstar-subtracted images are shown in Fig. 8 and Fig. 9. Wethen used the IRAF task ellipse to derive the integratedmagnitudes for all three galaxies; they are shown in Fig.10 and listed in Table 4 together with data from 2MASSand DENIS. We calibrated the AAT images by comparingthe AAT point sources, measured as described above, withsources from the 2MASS Point Source Catalog3 .

The radial surface brightness profiles are shown inFig. 11. As in Beijersbergen et al. (1999) and Galaz et al.(2002), we simultaneously fit the surface brightness data bytwo exponential profiles, taken to be the bulge and disk com-ponents of the galaxy:

µ(r) = µ + 1.086 ×r

hmag arcsec−2 (3)

The surface brightness profile parameters are given in Ta-ble 3. We used a geometric (multiplicative) step when mea-suring the surface brightness profile because such a step doesnot oversample the outer regions, where little flux is de-tected. Changing to a linear step gives more weight to the

2 IRAF is distributed by the National Optical Astronomy Obser-vatories, which are operated by the Association of Universities forResearch in Astronomy, Inc., under cooperative agreement withthe National Science Foundation.3 This publication makes use of data products from the Two Mi-cron All Sky Survey, which is a joint project of the Universityof Massachusetts and the Infrared Processing and Analysis Cen-ter/California Institute of Technology, funded by the NationalAeronautics and Space Administration and the National ScienceFoundation.

disk component and tends to brighten the disk central sur-face brightness by an average of 0.7 mag and decrease thedisk scale height by ∼1′′; the change in the bulge param-eters is less significant. A linear step also changes the cen-tral extinction-corrected (H-Ks) colours of the bulge andthe disk, from 0.41 and 0.50, to 0.44 and 0.34, respectively.When corrected for the inclination of the galaxy assumingno optical depth, the face-on central surface brightness ofthe disk and bulge components, µfo = µi − 2.5 log(cosi),becomes fainter by 1 mag. This correction is almost exactlycancelled by the K-band extinction correction.

The bulge of HIZOAJ0836–43 has a central observedsurface brightness of µ = 14.81 (15.70) mag arcsec−2 in theKs- (H -) band images and a scale length of h ∼ 2.5′′ or 1.7kpc. The disk of HIZOAJ0836–43 has a central observedsurface brightness of µ = 16.75 (17.74) mag arcsec−2 anda scale length of h ∼ 7′′ or 4.7 kpc. The bulge-to-disk ratio,B/D, calculated from the surface brightness profile fits, is0.80 (0.83) in the Ks- (H -) band.

The integrated infrared magnitudes are given in Table 4,together with the infrared properties measured in 2MASS,and DENIS. The formal statistical error on all AAT mag-nitudes listed in Table 4 is ∼0.01 mag; this error does notinclude potential systematic errors introduced during imageprocessing. The inclination of HIZOAJ0836–43, measuredfrom the surface brightness fits to the star-subtracted im-ages, i ∼ 66, was calculated assuming an intermediate valueof p = c/a = 0.15 (see Haynes & Giovanelli 1984).

4 THE GALAXY HIZOA J0836–43

The overall properties of the galaxy HIZOAJ0836–43, de-rived from our H i and infrared data, are summarised inTable 2. Below, we discuss the H i and dynamical mass ofHIZOAJ0836–43, along with its star formation rate, mor-phology, NIR colours, location on the Tully-Fisher relation,and environment.

4.1 The H i and dynamical mass

The H i flux density of HIZOAJ0836–43 as obtained fromthe Parkes HIZOA data is FHI = 13.2±1.3 Jy kms−1, similarto the value of 14.5±0.7 Jy km s−1 measured with the ATCAin 750D array (see Table 1), suggesting that little (if any)H i emission was resolved out by the interferometer in thisconfiguration. We adopt the ATCA value to estimate an H i

mass of MHI = 7.5 (±0.3) ×1010 M. Assuming the othertwo galaxies in the field (see Figs. 3 and 8) lie at the samedistance as HIZOAJ0836–43, we estimate an upper limit totheir individual H i masses of ∼ 5 × 109 M.

The H i diameter of HIZOAJ0836–43 is at least 3′, cor-responding to 120 kpc. Our analysis of the H i radial densityprofile indicates a face-on radius of 66 kpc, at a surface den-sity of 1 M pc−2. Using the H i radius and the inclination-corrected rotational velocity of vi

rot = 304 kms−1, we derivea total enclosed dynamical mass of Mdyn = 1.4 × 1012 M.The ratio of H i mass to dynamical mass is 0.053.

c© 0000 RAS, MNRAS 000, 000–000


Figure 8. Star-subtracted AAT Ks-band image (top) and H-band image (bottom) of HIZOA J0836–43 and neighbouring galaxies.

Figure 9. Star-subtracted AAT Ks-band images of 2MASX J08370723–4339137 (left), HIZOA J0836–43 (middle) and 2MASXJ08363600–4337556 (right). The contour levels are 30, 80, 160, 320, 640, 1280, 2560, 5120, and 10240 units. The images were convolvedwith a 1.′′5 Gaussian.

c© 0000 RAS, MNRAS 000, 000–000


Figure 10. Enclosed AAT Ks-band (top) and H-band (bottom) elliptical magnitudes as a function of semi-major axis radius (see alsoTable 4). Magnitudes have not been corrected for extinction.

4.2 Star formation rate

The 20-cm radio continuum data reveal a slightly extendedsource coincident with HIZOAJ0836–43 (see Fig. 3). Thesource has a total flux density of ∼24 mJy, a deconvolvedsize of 11′′×7′′, and a major axis position angle of 120. Thecentre position of the radio continuum source is α, δ(J2000)= 08:36:51.54, –34:37:41.5 (±0.1′′). Using the relation cal-ibrated by Bell (2003), we calculate a star formation rate(SFR) of 35 M yr−1. Assuming that Mgas = 1.4×MHI, thegas consumption time, Mgas/SFR, for HIZOAJ0836–43 is 3Gyr and is therefore quite typical for spiral galaxies (Ken-nicutt et al. 1994).

The far-infrared (FIR) 60 µm flux density of HI-ZOAJ0836–43 from the High Resolution Infrared Astro-nomical Satellite (IRAS) Atlas (Cao et al. 1997) confirmsthat star formation, and not AGN activity, is responsiblefor the radio continuum emission. Although HIZOAJ0836–43 is clearly detected at 60 µm (and marginally detected at100 µm), source confusion prevents measurement of an ac-curate flux. We instead place conservative upper and lowerlimits on the 60 µm flux density by summing the galaxyemission inside an elliptical aperture and subtracting boththe low and high extremes of the local sky; we measure a 60µm flux density of ∼ 1.5 − 4.8 Jy.

From the FIR 60 µm flux density limits, the ra-dio/infrared correlation for star-forming galaxies and radio-quiet AGN (Yun, Reddy, and Condon 2001) predicts radioluminosities that are a factor of 0.5 – 1.6 times that ob-

served. The observed radio luminosity therefore exceeds theexpected luminosity by at most a factor of 2, falling far be-low the factor of 5 required for ’radio-excess’ AGN (Yun etal. 2001, Drake et al. 2003), and confirming that the radiocontinuum emission is dominated by star formation.

While there is no sign of any radio continuum emissionat the position of the galaxy 2MASX J08363600–4337556,west of HIZOAJ0836–43, the smaller galaxy to the south-east of HIZOAJ0836–43, 2MASX J08370723–4339137, is de-tected with a 20-cm flux density of ∼1 mJy (see bottompanel of Fig 3). The rms in the continuum image is ∼0.1mJy beam−1.

4.3 Morphology and colours

The morphology of HIZOAJ0836–43 remains a puzzle. TheH i observations show a rapidly rotating, very gas-rich galaxywith — as seen in the previous section — a high star for-mation rate, indicative of an Sb or Sc galaxy (Roberts &Haynes 1994, Kennicutt 1992). In addition, the MHI/Mdyn

ratio of HIZOAJ0836–43 appears typical for Sb–Sc galax-ies, but would be within the observed range of values for anygalaxy type (Roberts & Haynes 1994). Visual examination ofthe AAT Ks-band image of HIZOAJ0836–43 (Fig. 3), how-ever, and the analysis of its surface brightness profile (Fig. 11and Table 3), reveal a distinct bright bulge (∼2 kpc) and anextended smooth disk (∼5 kpc) that can be traced along themajor axis out to a radius of ∼13.4 kpc (at 20 mag arcsec−2).Comparison with the NIR Galaxy Morphology Atlas of Jar-

c© 0000 RAS, MNRAS 000, 000–000


Figure 11. Observed AAT Ks- and H-band surface brightness (SB) profiles. The profiles of HIZOA J0836–43 and 2MASX J08363600–4337556 were simultaneously fit by two exponential profiles, given by the dot-dashed lines. The solid line shows the sum of the two fits.The profile of the third galaxy, 2MASX J08370723–4339137, was fit by a single exponential profile.

rett (2000) leads to the conclusion that HIZOAJ0836–43,with its dominant bulge and smooth bulge-to-disk transi-tion, is more likely to be a bright S0 to Sa galaxy than anSb or Sc. NIR extinction from the Galactic Plane, however,is likely to obscure the disk, causing HIZOAJ0836–43 toappear anomalously early in type. Using the NIR positionsof HIZOAJ0836–43 and its companions, we identify highlyobscured optical counterparts in the DSS2 R-band images,but the image resolution is insufficient to provide any detailsabout their morphologies.

The extinction-corrected NIR colours of HIZOAJ0836–43, presented in Table 5, are consistent with expectations fornormal bright galaxies (J −Ks = 1.m0, H −Ks = 0.m27, andJ−H = 0.m73), but provide little morphological informationas only late-type spirals (Sd – Sm) deviate from the overallmean (Huchra et al. 2000; Jarrett et al. 2003). The AAT(H − Ks)

o

5′′ values of 0.m39 and 0.m34 for HIZOAJ0836–43

and its larger companion, 2MASX J08363600–4337556, lieon the high side of the range for normal 2MASS galaxies; thesame colours determined from the 2MASS magnitudes lie onthe low side (0.m23 and 0.m15). We use the 5′′ colours insteadof the total colours because the former are less likely to beaffected by extinction and star-subtraction. The (J − Ks)colour of the small galaxy, 2MASX J08370723–4339137, lieswell above the mean, which might be indicative of a late type

spiral. In addition, comparison of DENIS central and mean(I−Ks) colours suggests that the total stellar populations ofHIZOAJ0836–43 and 2MASX J08363600–4337556 are bluerthan that of the bulges that dominate the light within 7′′.

As shown by Jarrett et al. (2003), NIR effective surfacebrightnesses brighten with decreasing morphological type,T . After extinction correction, the mean effective surfacebrightnesses listed in Table 2 suggest that HIZOAJ0836–43is of early type, in the range E (elliptical) to Sa, although theH and Ks surface brightnesses lie within the rms uncertain-ties for types as high as Sb. The face-on companion 2MASXJ08363600–4337556 also seems of early type, in the range Eto Sa, and the third small companion 2MASX J08370723–4339137 seems to be a spiral between Sab and Sc, in agree-ment with its S-shaped structure and the indication of starformation (see the bottom panel of Fig. 3).

As a final test of morphology, we compare the centralsurface brightness in the J- and Ks-band with the total Ks

magnitude. According to Jarrett (2000), SA0/S0 galaxiesfollow a linear relation of the form J5” = 0.8Ks + 8.8, andK5” = 0.8Ks + 7.8. Spiral galaxies of later type (T ) fol-low the same slope but with an increasing offset towardsfainter magnitudes with increasing T , while ellipticals gen-erally lie below this regression. The J5” and the K5” relationsagain suggests that the massive spiral galaxy HIZOAJ0836–

c© 0000 RAS, MNRAS 000, 000–000


43 is an S0 (lenticular) or Sa galaxy. The larger companion,2MASX J08363600–4337556, also seems of early type. Forthe small spiral-like galaxy, 2MASX J08370723–4339137, thetest however, also indicates an early type which it clearly isnot. This may be due to the fact that this galaxy is verysmall and hardly resolved; the above relation is valid onlyfor well resolved galaxies.

The NIR morphology, colours, and magnitudes there-fore indicate that HIZOAJ0836–43 may be an S0 or Sagalaxy, although we caution that extinction is likely to ob-scure the disk, causing the bulge to dominate the photome-try and therefore bias our results towards early morpholog-ical types. Given its enormous H i content, it may at firstseem contradictory that HIZOAJ0836–43 could be a lentic-ular galaxy. However, some lenticulars are known to containa large amount of neutral gas. Such galaxies include low sur-face brightness lenticulars, of which the edge-on S0 galaxyNGC 5084 (Gottesman & Hawarden 1986) is a typical ex-ample (see also O’Neil et al. 2004), as well as high surfacebrightness lenticulars (Eder et al. 1991). For many S0 galax-ies — in particular polar ring galaxies — the gas seems tobe due to recent accretion events. Other H i-rich lenticulars(e.g., NGC 5084), however, have H i disks that are alignedwith their optical counterparts as well as similar gas andstellar kinematics, suggesting that the H i is intrinsic in ori-gin.

4.4 NIR Tully-Fisher relation

In order to test whether HIZOAJ0836–43 lies on theTully-Fisher relation, we compare our data to the NIR(R, I,H,Ks) Tully–Fisher relations as calibrated by Macri(2001). These relations, calibrated for both total and isopho-tal magnitudes (20 mag/arcsec2), exhibit slopes with smalldispersions (0.m19 − 0.m22).

We list in Table 6 the absolute magnitudes estimatedfrom Macri’s Tully-Fisher relations (using the inclination-corrected 20% velocity width, log wi

20 = 2.82) and theextinction-corrected, absolute magnitudes obtained fromAAT, 2MASS and DENIS photometry4. The offsets betweenthe predicted and measured magnitudes, as well as the slopesand intercepts of the regressions M = a + b(log wi

20 − 2.5),are also listed. For all bands, HIZOAJ0836–43 lies on theNIR TF relation, never deviating by more than 1σ, sugges-tive of a normal galaxy at the extreme high-mass end of theNIR TF relation.

4.5 The environment

Does HIZOAJ0836–43 form part of a cluster or a group thatallowed it to grow to its current dimensions through the ac-cretion of or merging with satellite galaxies, or does it residein a low-density environment where it remained undisturbedby galaxy interactions? Unfortunately, the study of the en-vironment of HIZOAJ0836–43 is made difficult by its loca-tion behind the plane of the Milky Way and behind the Vela

4 The internal extinction was corrected according to Masters etal. (2003) using Aλ = cλ log(a/b), with cλ = 1.16, 0.39, 0.26 forI, J,Ks, respectively, where (a/b) is the major-to-minor axis ratio.

SNR. With optical extinctions of ∼10 magnitudes, the detec-tion of even the intrinsically most luminous galaxies at thedistance of HIZOAJ0836–43 is nearly impossible. AlthoughNIR surveys are less affected by extinction than optical sur-veys, they are hindered by confusion. Galaxy recognitionfails completely in regions where the density of stars withK ≤ 14.m0 surpasses log N = 4.00(2), and HIZOAJ0836–43 (with log N = 3.96) lies extremely close to this so-called“NIR Zone of Avoidance”(see Kraan-Korteweg 2005).

Nevertheless, we attempt to explore the environmentof HIZOAJ0836–43 by studying all extended sources inthe 2MASX catalogue that lie within (1) a square degreearound HIZOAJ0836–43 and (2) within a circle of radius1.2, the equivalent of an Abell radius at the distance ofHIZOAJ0836–43. With the exception of two small pocketsof extreme dust, the extinction AK in this area ranges from0.m6 to 1.m4. Subtracting the latter from the nominal 2MASScompleteness limit of Ks ≤ 13.m5 (Jarrett 2000), we expectour extended source search to be complete to K

S<∼ 12.m1 in

extinction-corrected magnitude, ignoring confusion. To ex-clude Galactic sources from this sample, we visually inspectall 2MASX objects brighter than K

S ≤ 12.m0.

In the central square degree, there are 14 2MASX ex-tended sources with K

S ≤ 12.m0. Of these, one was re-jected, leaving 10 certain galaxies and 3 possible galaxies.The galaxy density of the studied region is therefore at leasta factor of 5 higher than the mean density of 2MASS galaxieswith KS ≤ 12.m0 and |b| ≥ 25, 1.8 gal/(2) (Jarrett 2004;Jarrett 2005, priv. comm.), although the redshift of the over-density remains undetermined. Within the Abell radius, 29extended sources have K

S ≤ 12.m0; of these, six were re-jected, leaving 18 certain galaxies plus 5 possible galaxies.The resulting galaxy density in the Abell radius thereforealso exceeds the mean, by a minimum factor of 2–3.

To investigate the effect of confusion on the overdensity,we repeated the study for completeness limits of K

S ≤ 11.m5and 11.m0. The mean 2MASX densities (away from theGalactic Plane) are 0.85 and 0.43 gal/(2), respectively. Al-though the statistics are poorer, we find galaxy densities inthe square degree around HIZOAJ0836–43 of 7−10 and 5−8gal/(2), respectively, indicating even higher overdensitiesthan for the K

S ≤ 12.m0 completeness limit. The overdensitywithin the Abell radius remains approximately the same. Ittherefore appears that HIZOAJ0836–43 lies in an overdenseregion — at least projected on the sky — with the galaxydensity increasing in the central square degree and becomingmore prominent for the brightest objects.

Although we do not have redshifts for the individual2MASS galaxies, we estimate the redshift of the deducedgalaxy overdensity, and its relation to the redshift of HI-ZOAJ0836–43, z = 0.036, by considering the large-scalestructure of 2MASS galaxies in several photometric redshiftintervals, as determined by Jarrett (2004). Prominent in theslice 0.04 < z < 0.05, but also visible in the 0.03 < 0.04slice, is a filamentary supercluster-like feature that crossesthe Galactic Plane at the position of HIZOAJ0836–43 andthat points towards the Shapley concentration. This featurecould explain the projected overdensity of galaxies aroundHIZOAJ0836–43. In addition, a large fraction of the 2MASSgalaxies seem to be of similar size and appearance as HI-ZOAJ0836–43 (taking absorption effects into account), andare therefore likely to lie at similar redshifts.

c© 0000 RAS, MNRAS 000, 000–000


Finally, we checked for H i-rich neighbours to HI-ZOAJ0836–43 by observing a 2 × 2 area around it withthe Parkes multibeam system in 2004 November. The in-tegration time was 4× longer than for the deep Parkes H i

ZOA survey, and 20× longer than for HIPASS. The centralvelocity was set to 10 000 km s−1; the 64 MHz bandwidthgive velocity coverage from 3300 kms−1 to 16 700 kms−1. Acareful inspection of the deep H i data cube did not revealany additional galaxies. Given an rms of >

∼ 3 mJy, the 5σ H i-

mass limit for a detection at D = 148 Mpc is >∼ 1.5×1010M

(assuming a top-hat with a velocity width of 200 kms−1).Areas with strong continuum emission are adversely affectedby a baseline ripple which increases the noise by a factor 2–3.

5 COMPARISON WITH OTHER H i-MASSIVE

GALAXIES

The four most H i-massive galaxies known are HI-ZOAJ0836–43, UGC 4288 (O’Neil et al. 2004), UGC 1752(Matthews et al. 2001), and, of course, Malin 1 (Bothun etal. 1987). The latter three are low-surface brightness (LSB)galaxies at luminosity distances of 433, 249 and 351 Mpc,respectively. Their H i properties are summarised in Table 7.

The galaxy UGC 4288 was classified as Sdm bySchombert & Bothun (1988) and as S0 by O’Neil et al.(2004). Optical images show a bright bulge and faint spiralstructure, suggesting that UGC 4288 is likely to be of latetype. The galaxy UGC 1752, which was classified as SAcdby de Vaucouleurs et al. (1991), looks remarkably similar toUGC 4288, suggesting the latter is also of late type. Malin 1has a prominent bulge, with the properties of a bright ellip-tical galaxy, and a low-surface brightness disk that is barelydetectable. The galaxy HIZOAJ0836–43 is highly obscuredin the optical and its morphological type is unknown. Whileour analysis of its NIR colours slightly favour an early-typemorphology, extinction is likely to obscure the disk, causingthe measurements to be dominated by the prominent bulge.The surface brightness analysis gives B/D ∼ 0.8, a valueslightly higher than that seen in LSBs (B/D = 0.56 ± 0.07;Sprayberry et al. 1995), but roughly consistent with thatof spiral galaxies (B/D = 0.61 ± 0.20; Kent 1985). Theextinction- and inclination-corrected central Ks-band sur-face brightness of the disk (µ=16.9 mag arcsec−2) falls wellshort of the typical LSB cut, 18.0 mag arcsec−2, indicatingthat unlike the other H i-massive galaxies listed above, HI-ZOAJ0836–43 is not an LSB galaxy.

The H i diameters of all four galaxies are huge: Malin 1(∼2.′5 or 220 kpc), UGC 1752 (∼2′ or 130 kpc), and HI-ZOAJ0836–43 (∼3′ or 120 kpc). No H i diameter has beenmeasured for UGC 4288 (yet) but its optical extent of ∼1′

(or 100 kpc) is likely to be a lower limit. All four galaxiesalso have high rotational velocities; while their inclinationangles are rather uncertain (Malin 1: i ≈ 45, UGC 1752:i ≈ 0; UGC 4288 i ≈ 40; HIZOAJ0836–43: i ≈ 66), wededuce rotational velocities of >

∼ 200–300 kms−1 from thevelocity width of their H i profiles (see Table 7). We note thatUGC 4288, UGC 1752, and Malin 1 appear to lie in very lowdensity environments. Their lack of tidal interactions mayhave played a role in preserving such large H i masses anddiameters.

It therefore appears that the four H i-massive galaxies

discussed above are similar in their extreme H i masses, largedisks, and high rotational velocities. It is possible, however,that HIZOAJ0836–43 differs from the others in its highSFR, high surface brightness disk, potentially overdense en-vironment, and morphology. A more detailed comparison ofthese galaxies will follow in Koribalski et al. (in prepara-tion).

6 CONCLUSIONS

HIZOA J0836–43 is a remarkable galaxy. Its enormous H i

mass (MHI = 7.5 (±0.3) ×1010 M), large H i velocity width(w20 = 610± 9 kms−1), and paucity of optical informationdue to its location in the Zone of Avoidance raises numerousquestions about its environment, morphology and overallkinematics. ATCA H i observations reveal a very extended,highly-inclined H i disk (see Fig. 2) and a surprisingly regu-lar velocity field. We measure an H i diameter of at least 3′

which corresponds to a physical size of ∼120 kpc. For com-parison, the H i diameter of Malin 1, the largest disk galaxyknown so far, is ∼220 kpc (Pickering et al. 1997, Matthewset al. 2001).

The H i distribution of HIZOAJ0836–43 is slightlyasymmetric (with more H i gas on the eastern side of thegalaxy) and reveals a central depression. It also shows agradually changing major axis position angle which de-creases from PA ∼ 290 in the inner region to nearly 270

in the outermost region. This change, which is most pro-nounced on the eastern side of the galaxy, is indicative of agentle warp.

The H i velocities observed in HIZOAJ0836–43 rangefrom ∼ 10400 to 11000 kms−1, indicating an inclination-corrected rotation velocity of 304 kms−1 and consequently,a large dynamical mass and dark matter content. HI-ZOAJ0836–43 falls on the NIR Tully-Fisher relationship,indicating that it is a normal rotating galaxy. We note thatthe equal velocity contours are nearly straight and do not re-semble a typical spider diagram. A detailed analysis of thegalaxy kinematics is limited by the relatively low angularresolution.

NIR data from the AAT indicate that HIZOAJ0836–43 has a both a prominent bulge and an extended disk,and NIR colours point to an early type (S0-Sa) morphol-ogy, although colours from AAT, 2MASS, and DENIS maybe dominated by the bulge component due to obscurationof the disk. The high star formation rate measured fromthe radio continuum and IRAS data, SFR ∼ 35 M yr−1,suggests a later morphological type of Sb or Sc. A studyof the local environment of HIZOAJ0836–43, using sourcesfrom the 2MASX extended source catalogue, indicate thatHIZOAJ0836–43 may lie in an overdense region, possiblypart of a supercluster-like filament.

There are currently very few galaxies known with H i

masses of ∼ 1011M; these are HIZOAJ0836–43, UGC 1752,Malin 1, and UGC 4288 (in order of increasing distance; seeTable 7). Accurate H i mass estimates can be difficult forsuch distant galaxies because of flux calibration uncertain-ties and low signal to noise. In addition, single-dish H i spec-tra are often fitted with multiple-order baselines which resultin large flux uncertainties due to the very wide H i velocityprofiles of these supermassive galaxies. We have obtained

c© 0000 RAS, MNRAS 000, 000–000


deep H i spectra of the LSB galaxies Malin 1 and UGC 4288with the Parkes 64-m telescope and will report the results inKoribalski et al. (in preparation) together with a comprehen-sive comparison of the most H i-massive galaxies known fromthe literature. Constraining the number of massive galaxiesin the Universe is necessary if we are to better understandthe high-mass end of the galaxy mass and luminosity func-tions, and studying the properties of such galaxies providesa test of current galaxy formation scenarios in their mostextreme limits.

ACKNOWLEDGMENTS

We are grateful to Patrick Woudt for his help with thestar-subtraction in the AAT K- and H-band images. Wethank Stacy Mader for obtaining the sensitive Parkes H i

data of the 2 × 2 area around HIZOAJ0836–43, andthe referee, Greg Bothun, for helpful comments. This re-search has made use of the NASA/IPAC Infrared Sci-ence Archive (2MASS) and the NASA/IPAC ExtragalacticDatabase (NED), which are operated by the Jet Propul-sion Laboratory, California Institute of Technology, undercontract with the National Aeronautics and Space Admin-istration. RCKK thanks UCT, the SA National ResearchFoundation and CONACyT (research grant 40094F) fortheir support, and the Australian Telescope National Facil-ity (ATNF, CSIRO) for their hospitality during her sabbat-ical. JLD performed most of this research while a Fulbrightscholar at the ATNF.

REFERENCES

Barnes, D.G. et al., 2001, MNRAS, 322, 486Beijersbergen, M., de Blok, W.J.G., van der Hulst, J.M., 1999,

A&A, 351, 903Bell, E.F., 2003, ApJ, 586, 794Blanton, M.R. et al., 2003, ApJ, 592, 819Bothun, G.D., Impey, C.D., Malin, D.F., Mould, J.R., 1987, AJ,

94, 23

Cao, Y., Terebey, S., Prince, T.A., Beichman, C.A., 1997, ApJS,111, 387

Cole, S., Lacey, C.G., Baugh, C.M., Frenk, C.S., 2000, MNRAS,319, 168

Colina, L., Bohlin, R., 1997, AJ, 113, 1138Croton, D.J. et al., 2006, MNRAS, 365, 11

Davies, J.I., Sabatini, S., Roberts, S., 2004, PASA, 21, 360Donley, J.L. et al., 2005, AJ, 129, 220Drake, C.L., McGregor, P.J., Dopita, M.A., van Breugel, W.J.M.,

2003, AJ, 126, 2237Eder, J., Giovanelli, R., Haynes, M.P., 1991, AJ, 102, 572Epchtein, N., 1997, in Garzon, F. et al., eds, 2nd DENIS Euro-

conference, The impact of large scale near-infrared surveys.

Dordrecht, Kluwer, 15Fixsen, D.J., Cheng, E.S., Gales, J.M., Mather, J.C., Shafer, R.A.,

Wright, E.L., 1996, ApJ, 473, 576Foerster Schreiber, N.M. et al., 2004, ApJ, 616, 40Galaz, G., Dalcanton, J.J., Infante, L., Treister, E., 2002, AJ, 124,

1360Gooch, R., 1996, in Jacoby, G.H., Barnes,J., eds, ASP Conf. Ser.

Vol. 101, Astronomical Data Analysis Software and SystemsV. Astron. Soc. Pac., San Francisco, p. 80

Gottesman, S.T., Hawarden, T.G., 1986, MNRAS, 219, 759Haynes, M.P, Giovanelli, R., 1984, AJ, 89, 758

Henning, P.A. et al., 2000, AJ, 119, 2686

Henning, P.A., Kraan-Korteweg, R.C., Staveley-Smith, L., 2005,in Fairall, A.P., Woudt, P.A., eds, ASP Conf. Ser. Vol. 329,Nearby Large-Scale Structures and the Zone of Avoidance.Astron. Soc. Pac., San Francisco, p. 199

Huchra, J.P., Mader, J., Schneider, S.E., Skrutskie, M.F., Jar-rett, T.H., Chester, T., Cutri, R., 2000, in Kraan-Korteweg,R.C., Henning, P.A., Andernach, H., eds, ASP Conf. Ser. Vol.218, Mapping the Hidden Universe. Astron. Soc. Pac., SanFrancisco, p. 101

Impey, C.D., Bothun, G., 1989, ApJ, 341, 89Jarrett, T.H., 2000, PASP, 112, 1008

Jarrett, T.H., 2004, PASA, 21, 396Jarrett, T.H., Chester, T., Cutri, R., Schneider, S., Skrutskie, M.,

Huchra, J.P., 2000, AJ, 119, 2498

Jarrett, T.H., Chester, T., Cutri, R., Schneider, S.E., Huchra,J.P., 2003, AJ, 125, 525

Kauffmann, G., White, S.D.M., Guiderdoni, B., 1993, MNRAS,264, 201

Kay, S.T., Pearce, F.R, Frenk, C.S., Jenkins, A., 2002, MNRAS,330, 113

Kennicutt, R.C., 1992, ApJ, 388, 310

Kennicutt, R.C., Tamblyn, P., Congdon, C.E., 1994, ApJ, 435, 22Kent, S. M. 1985, ApJS, 59, 115

Klypin, A., Nolthenius, R., Primack, J., 1997, ApJ, 474, 533

Koribalski, B.S. et al., 2004, AJ, 128, 16Kraan-Korteweg, R.C., 2005, in Roeser, S., ed, Rev. in Modern

Astrophysics Vol. 18, From Cosmological Structures to theMilky Way. p. 48

Kraan-Korteweg, R.C., Staveley-Smith, L., Donley, J.L., Koribal-ski, B.S., Henning, P.A., 2004, in Colless, M., Staveley-Smith,L., Stathakis, R. eds, IAU Symp. 216, Maps of the Cosmos,Astron. Soc. Pac. San Franciso, p. 203

Lucy, L.B., 1974, AJ, 79, 745Macri, L.M., 2001, PhD Thesis, Harvard University

Malhotra, S. 1994, ApJ, 433, 687Masters, K.L., Giovanelli, R., Haynes, M.P., 2003, AJ, 126, 158

Matthews, L.D., van Driel, W., Monnier-Ragaigne, D., 2001,A&A, 365, 1

Meyer, M.J. et al., 2004, MNRAS, 350, 1195Norberg, P. et al., 2002, MNRAS, 336, 907

Olling, R.P., 1996, AJ, 112, 457

O’Neil, K., Bothun, G., van Driel, W., Monnier Ragaigne, D.,2004, A&A, 428, 823

Pickering, T.E., Impey, C.D., van Gorkom, J.H., Bothun, G.D.,1997, AJ, 114, 1858

Press, W.H., Schechter, P., 1974, ApJ, 187, 425

Press, W.H., Teukolsky, S.A., Vetterling, W.T., Flannery, B.P.,1992, Numerical recipes in C : the art of scientific computing,2nd edition. Cambridge University Press, Cambridge

Roberts, M.S., Haynes, M.P., 1994, ARA&A, 32, 115

Rosenberg, J.L., Schneider, S.E., 2002, ApJ, 567, 247Schechter, P., 1976, ApJ, 203, 297

Schlegel, D.J., Finkbeiner, D.P., Davis, M., 1998, ApJ, 500, 525

Schombert, J.M., Bothun, G.D., 1988, AJ, 95, 1389Sofue, Y., 1996, ApJ, 458, 120

Sofue, Y., 1997, PASJ, 49, 17

Sprayberry, D., Impey, C. D., Bothun, G. D., Irwin, M. J., 1995,AJ, 109, 558

Springel, V. et al., 2005, Nature, 435, 629

Stark, A.A., Brand, J., 1989, ApJ, 339, 763

Staveley-Smith, L. et al., 1996, PASA, 13, 243van der Hulst, J.M., Terlouw, J.P., Begeman, K.G., Zwitser, W.,

Roelfsema, P.R., 1992, in Worall, D.M., Biemesderfer, C.,Barnes, J., eds, ASP Conf. Ser. Vol. 25, Astronomical DataAnalysis Software and Systems I. Astron. Soc. Pac., San Fran-cisco, p. 131

de Vaucouleurs, G., de Vaucouleurs, A., Corwin Jr., H.G., Buta,

c© 0000 RAS, MNRAS 000, 000–000


R.J., Paturel, G., Fouque, P., 1991, Third Reference Cata-

logue of Bright Galaxies. Springer Verlag, New YorkWarmels, R.H., 1988, A&AS, 72, 427Wong, I., et al., 2005, MNRAS, submittedWoudt, P.A., Fairall, A.P., Kraan-Korteweg, R.C., Lucey, J.,

Schroder, A., Burstein, D., McCall, M.L., 2005, in Fairall,A.P., Woudt, P.A., eds, ASP Conf. Ser. Vol. 329, NearbyLarge-Scale Structures and the Zone of Avoidance. Astron.Soc. Pac., San Francisco, p. 147

Yun, M.S., Reddy, N.A., Condon, J.J., 2001, ApJ, 554, 803Zwaan, M.A. et al., 2003, AJ, 125, 2842Zwaan, M.A., Meyer, M.J., Staveley-Smith, L., Webster, R.L.,

2005, MNRAS, 359, L30

c© 0000 RAS, MNRAS 000, 000–000


Table 1. Measured H i properties of the galaxy HIZOA J0836–43

Observation rms H i flux density Heliocentric Velocity width ReferenceFHI velocity, cz w50 w20

[mJy beam−1] [Jy km s−1] [ km s−1] [ km s−1] [ km s−1]

Parkes ZOA survey 6 13.2 ± 1.3 10706 ± 7 550 ± 14 607 ± 21 hereATCA 1.2 14.5 ± 0.7 10689 ± 3 566 ± 6 610 ± 9 hereHIPASS 13 15.6 ± 2.9 10710 ± 16 575 ± 32 654 ± 48 HICAT (Meyer et al. 2004)

Note: The uncertainties were calculated following the formulas given in Koribalski et al. (2004) and references therein.

Table 2. A summary of the properties of the galaxy HIZOA J0836–43

Property Value Notes

centre position, α, δ(J2000) 08h36m51.54s, –4337′41.′′5 (a)heliocentric velocity, vhel 10689 ± 3 km s−1 (b)CMB velocity, vCMB 10928 ± 3 km s−1 · · ·

luminosity distance 148 Mpc (c)H i flux density, FHI 14.5 ± 0.7 Jy km s−1 (b)H i mass, MHI 7.5 × 1010 M (b)position angle, PA 285 ± 15 (d)inclination angle, i 66 ± 3 (d)intrinsic rotation velocity, vi

rot 304 km s−1 · · ·

H i radius (at 1 M pc−2) 66 kpc (e)total dynamical mass, Mdyn 1.4 × 1012 M · · ·

absolute magnitude, McKs

–25.03 mag (d,f,g)

infrared colour, (H − Ks) 0.43 (d,f,g)

H i mass to light ratio, MHI / LKs0.35 M/L (f,g)

H i mass to total mass ratio, MHI / Mdyn 0.05 · · ·

infrared radius, (µcKs

= 20 mag arcsec−2) 13.4 kpc (d,f)

bulge scale length 1.7 kpc (d)disk scale length 4.7 kpc (d)star formation rate (SFR) 35 M yr−1 (a)

Notes: (a) from the ATCA 20-cm radio continuum data; (b) from the ATCA 750D array H i line data; (c) assuming H0 =75 km s−1 Mpc−1 (see § 3.1.1); (d) from the AAT H- and Ks-band data; (e) when viewed face-on; (f) corrected for Galactic extinction(Schlegel et al. 1998); (g) the absolute magnitude of the Sun in K-band is MK, = 3.31 mag (Colina & Bohlin 1997).

Table 3. Observed Surface Brightness Profile Parameters

Bulge Parameters Disk ParametersSource Band µ µ h µ h

[mag arcsec−2] [mag arcsec−2] [′′] [mag arcsec−2 ] [′′]

HIZOA J0836–43 H 15.55 15.70 2.53 17.74 7.15Ks 14.65 14.81 2.48 16.75 6.79

2MASX J08363600–4337556 H 15.20 15.27 1.34 18.29 6.25Ks 14.27 14.37 1.16 16.93 5.40

2MASX J08370723–4339137 Ks 15.87 15.87 1.81 · · · · · ·

Note: Magnitudes have not been corrected for foreground extinction, values of which are listed in Table 4.

c© 0000 RAS, MNRAS 000, 000–000


Table 4. Observed Infrared Parameters

Source Band Aλ Telescope r20 reff mtot m20 m(r<5′′) SB(reff)[mag] [′′] [′′] [mag arcsec−2 ]

HIZOA J0836–43 I 4.40 DENIS · · · · · · 16.59 ± 0.10 · · · · · · · · ·

J 2.05 2MASS 11.2 7.7 13.16 ± 0.13 13.25 ± 0.08 13.99 ± 0.05 18.75· · · DENIS · · · · · · 13.17 ± 0.04 · · · · · · · · ·

H 1.31 AAT 16.1 6.0 11.73 ± 0.01 11.84 ± 0.01 12.39 ± 0.01 17.77· · · 2MASS 20.3 10.8 11.44 ± 0.06 11.81 ± 0.06 12.55 ± 0.03 17.71

Ks 0.83 AAT 20.2 6.3 10.82 ± 0.01 10.90 ± 0.01 11.52 ± 0.01 17.02· · · 2MASS 17.8 9.6 10.84 ± 0.06 11.18 ± 0.06 11.84 ± 0.02 16.86· · · DENIS · · · · · · 11.02 ± 0.10 · · · · · · · · ·

2MASX J08363600–4337556 I 4.27 DENIS · · · · · · 17.04 ± 0.08 · · · · · · · · ·

J 1.99 2MASS 6.7 5.1 13.37 ± 0.14 13.43 ± 0.10 14.10 ± 0.05 18.81· · · DENIS · · · · · · 13.63 ± 0.03 · · · · · · · · ·

H 1.27 AAT 11.2 4.5 12.00 ± 0.01 12.22 ± 0.01 12.62 ± 0.01 18.22· · · 2MASS 11.6 6.6 11.84 ± 0.08 12.08 ± 0.08 12.76 ± 0.04 17.84

Ks 0.81 AAT 15.6 5.4 11.08 ± 0.01 11.18 ± 0.01 11.82 ± 0.01 17.76· · · 2MASS 13.9 6.0 11.31 ± 0.08 11.45 ± 0.08 12.15 ± 0.03 17.06· · · DENIS · · · · · · 11.41 ± 0.06 · · · · · · · · ·

2MASX J08370723–4339137 J 1.95 2MASS · · · 2.5 15.92 ± 0.37 · · · 16.33 ± 0.40 19.91H 1.24 2MASS · · · 2.2 14.55 ± 0.25 · · · 14.62 ± 0.19 18.24Ks 0.79 AAT 7.3 2.7 13.13 ± 0.01 13.24 ± 0.01 13.29 ± 0.01 17.50

· · · 2MASS 5.4 3.0 13.51 ± 0.22 13.71 ± 0.14 13.77 ± 0.14 17.89

Table columns: (1) Source name; (2) NIR band; (3) extinction within the given band; (4) telescope; (5) r20 is the radius at 20 magarcsec−2; (6) reff is the integrated half-light radius; (7) mtot is the total apparent magnitude; (8) m20 is the 20 mag arcsec−2 ellipticalisophotal magnitude (9) m(r<5′′) is the 5′′ circular aperture magnitude; and (10) SB(reff) is the mean surface brightness at reff .

Table 5. NIR colours corrected for galactic foreground extinction

2MASX 2MASXColour HIZOA J0836–43 J08363600–4337556 J08370723–4339137

AAT 2MASS DENIS AAT 2MASS DENIS 2MASS

(I − J)oT · · · · · · 1.07 · · · · · · 1.13 · · ·

(I − J)o

7′′ · · · · · · 0.64 · · · · · · 0.76 · · ·

· · · · · · · · · · · · · · · · · · · · ·

(I − Ks)oT · · · · · · 2.00 · · · · · · 2.17 · · ·

(I − Ks)o

7′′ · · · · · · 1.50 · · · · · · 1.49 · · ·

· · · · · · · · · · · · · · · · · · · · ·

(J − Ks)oT · · · 1.10 0.93 · · · 0.88 1.04 1.25

(J − Ks)o

5′′ · · · 0.93 0.87∗ · · · 0.77 0.72∗ 1.40

· · · · · · · · · · · · · · · · · · · · ·

(H − Ks)oT 0.43 0.12 · · · 0.46 0.07 · · · 0.59

(H − Ks)o

5′′ 0.39 0.23 · · · 0.34 0.15 · · · 0.40

· · · · · · · · · · · · · · · · · · · · ·

(J − H)oT · · · 0.98 · · · · · · 0.81 · · · 0.66

(J − H)o

5′′ · · · 0.70 · · · · · · 0.62 · · · 1.00

Note: ∗measured at 7′′.

c© 0000 RAS, MNRAS 000, 000–000


Table 6. NIR Tully-Fisher comparisons

Mag Intercepta Slopea Mλ(TF) Mλ(phot) ∆(TF-phot)

IT −21.10 −8.7 −23.88 −23.92 (DE) +0.04 (DE)

HT −22.40 −10.0 −25.60 −25.52 (AAT) −0.08 (AAT)−25.81 (2M) +0.21 (2M)

KT −22.66 −9.9 −25.82 −26.04 (AAT) +0.22 (AAT)−25.90 (2M) +0.08 (2M)−25.72 (DE) −0.10 (DE)

H20 −22.01 −10.5 −25.37 −25.41 (AAT) +0.04 (AAT)−25.44 (2M) +0.07 (2M)

K20 −22.34 −10.4 −25.67 −25.84 (AAT) +0.17 (AAT)−25.56 (2M) −0.11 (2M)

Notes: (a) Intercept and slope were taken from Macri 2001.

Table 7. H i measurements and derived properties of some of the most H i-massive galaxies known.

H i source vhel D FHI log MHI w50 w20 Ref., comments[ km s−1] [Mpc] Jy km s−1 [M]

UGC 4288 30223 433 2.54 11.05 520 558 1, AreciboUGC 1752 17861 249 5.13 10.87 388 429 2, NancayHIZOA J0836–43 10689 148 14.5 10.87 566 610 3, ATCA image

Malin 1 24784 351 1.8 10.72 293 341 2, Nancay” 24755 ” 2.5 10.86 322 4, VLA image” 24705 ” 2.7 10.89 295 355 5, Arecibo” 24745 ” 4.6 11.13 315 340 6, NRAO 43-m

References: (1) O’Neil et al. 2004, (2) Matthews et al. 2001, (3) this paper, (4) Pickering et al. 1997, (5) Bothun et al. 1987, and (6)Impey & Bothun 1989. Notes: The heliocentric velocity, vhel, given in Col.(2), was used to calculate the luminosity distance, D, givenin Col.(3) as described in Section 4.

c© 0000 RAS, MNRAS 000, 000–000

a massive spiral galaxy in the zone of avoidancemensa.ast.uct.ac.za/~kraan/papers/06massive.pdf ·...

Documents